Maintenance of information over a short period of time in visual working memory is a critical function for natural vision and various cognitive functions. However, working memory can be selectively impaired in many psychiatric (e.g., schizophrenia and depression) and neurological (e.g., Autism, ADHD, and Amnesic) conditions. The proposed research focuses on a better characterization of information stored in VWM by integrating both measures of quantity and quality in healthy participants, and by factoring out selective impairments in item and associative information on amnesic and schizophrenia patients. Working memory is typically assumed to differ from long-term memory in the sense that it reflects a limited capacity system consisting of a limited number of items. However, recent work has suggested that visual working memory may be able to store a rich amount of information from visual scenes and thus the capacity of working memory may be virtually unlimited. The proposed research aims to contrast these two opposing views by using quantitative psychophysical methods borrowed from research on long-term memory. Quantitative models that underlie these two opposing views will be tested against a newly proposed model that independently measures the quantity and quality of working memory representations. This approach will then be applied to amnesic and schizophrenia patients to address another major open issue in working memory. Specifically, are simple features (e.g., the shape of a strawberry) and relational information (e.g., the location and color of a strawberry in a box) in working memory supported by different mechanisms? Both research aims of this proposal are grounded on and integrated with two training goals. The first is to apply the theoretical constructs and quantitative methods from long-term memory studies to the proposed study on working memory. The second is to receive training in neuropsychological studies of memory and to apply that to the proposed work in working memory. This fellowship is vital for the long-term career development of the applicant, who seeks to develop an independent research program that combines quantitative psychophysics and translational approaches to study normal and abnormal memory functioning. PUBLIC HEALTH RELEVANCE: Maintenance of information over a short period of time in visual working memory not only supports natural vision (e.g., maintaining saccade targets across eye movements) and various cognitive functions (e.g., language comprehension and general intelligence), but also suffers severe impairment in many psychiatric (e.g., schizophrenia) and neurological (e.g., autism and ADHD) conditions. The proposed research assesses a novel way of characterizing working memory by integrating measures of two different aspects of working memory--quantity and quality--which have been the foundations for two major opposing hypotheses about limitations in working memory. The importance of this approach is made concrete by applying the proposed model to explore the possible impairment in object and associative information in working memory in amnesic patients who have suffered medial temporal lobe damage and schizophrenia patients.